https://bugs.exim.org/show_bug.cgi?id=2480
Bug ID: 2480
Summary: Heap buffer overflow vulnerability in GETCHARLEN()
(pcre2_match.c)
Product: PCRE
Version: 10.34 (PCRE2)
Hardware: x86
OS: All
Status: NEW
Severity: security
Priority: medium
Component: Code
Assignee: ph10@???
Reporter: clickwithsk@???
CC: pcre-dev@???
Created attachment 1235
-->
https://bugs.exim.org/attachment.cgi?id=1235&action=edit
Crash input file that triggers the bug
The latest version of PCRE (pcre2-10.34-RC1, pcre2-10.33) is prone to a Heap
Overflow vulnerability which can be triggered using a crafted regular
expression. It could allow a remote attacker to obtain sensitive information,
or cause denial of service. This vulnerability occurs because the library fails
to perform adequate boundary-checks on user-supplied data in GETCHARLEN()
(pcre2_match.c), which leads to a heap buffer overflow, causing an out of
bounds memory read, and subsequently resulting into a crash.
====================
Output of ASAN compiled library (-fsanitize=address)
Run as: ./pcre2test crashes/crash_file_2
---------------------
==35471==ERROR: AddressSanitizer: heap-buffer-overflow on address
0x629000004200 at pc 0x55555560239f bp 0x7fffffff7bb0 sp 0x7fffffff7ba0
READ of size 1 at 0x629000004200 thread T0
#0 0x55555560239e in match src/pcre2_match.c:1594
#1 0x55555560239e in pcre2_match_8 src/pcre2_match.c:6968
#2 0x555555576c41 in process_data src/pcre2test.c:7537
#3 0x55555557cd61 in main src/pcre2test.c:9004
#4 0x7ffff681db96 in __libc_start_main
(/lib/x86_64-linux-gnu/libc.so.6+0x21b96)
#5 0x55555557dbf9 in _start
(pcre-10.34-RC1/pcre2test-addrsan-fno-omit-gdwarf-g3+0x29bf9)
0x629000004200 is located 0 bytes to the right of 16384-byte region
[0x629000000200,0x629000004200)
allocated by thread T0 here:
#0 0x7ffff6ef8f40 in realloc
(/usr/lib/x86_64-linux-gnu/libasan.so.4+0xdef40)
#1 0x555555574899 in process_data src/pcre2test.c:6652
#2 0x55555557cd61 in main src/pcre2test.c:9004
#3 0x7ffff681db96 in __libc_start_main
(/lib/x86_64-linux-gnu/libc.so.6+0x21b96)
SUMMARY: AddressSanitizer: heap-buffer-overflow src/pcre2_match.c:1594 in match
Shadow bytes around the buggy address:
0x0c527fff87f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0x0c527fff8800: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0x0c527fff8810: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0x0c527fff8820: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0x0c527fff8830: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
=>0x0c527fff8840:[fa]fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
0x0c527fff8850: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
0x0c527fff8860: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
0x0c527fff8870: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
0x0c527fff8880: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
0x0c527fff8890: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
Shadow byte legend (one shadow byte represents 8 application bytes):
Addressable: 00
Partially addressable: 01 02 03 04 05 06 07
Heap left redzone: fa
Freed heap region: fd
Stack left redzone: f1
Stack mid redzone: f2
Stack right redzone: f3
Stack after return: f5
Stack use after scope: f8
Global redzone: f9
Global init order: f6
Poisoned by user: f7
Container overflow: fc
Array cookie: ac
Intra object redzone: bb
ASan internal: fe
Left alloca redzone: ca
Right alloca redzone: cb
==35471==ABORTING
=====================
valgrind output:
valgrind -v --leak-check=full ./pcre2test $crash_file_2
--------------------
==36076== HEAP SUMMARY:
==36076== in use at exit: 117,819 bytes in 14 blocks
==36076== total heap usage: 27 allocs, 13 frees, 124,674 bytes allocated
==36076==
==36076== Searching for pointers to 14 not-freed blocks
==36076== Checked 75,536 bytes
==36076==
==36076== LEAK SUMMARY:
==36076== definitely lost: 0 bytes in 0 blocks
==36076== indirectly lost: 0 bytes in 0 blocks
==36076== possibly lost: 0 bytes in 0 blocks
==36076== still reachable: 117,819 bytes in 14 blocks
==36076== suppressed: 0 bytes in 0 blocks
==36076== Reachable blocks (those to which a pointer was found) are not shown.
==36076== To see them, rerun with: --leak-check=full --show-leak-kinds=all
==36076==
==36076== ERROR SUMMARY: 1 errors from 1 contexts (suppressed: 0 from 0)
==36076==
==36076== 1 errors in context 1 of 1:
==36076== Invalid read of size 1
==36076== at 0x13F175: pcre2_match_8 (in pcre-10.34-RC1/pcre2test)
==36076== by 0x111C49: process_data (in pcre-10.34-RC1/pcre2test)
==36076== by 0x115703: main (in pcre-10.34-RC1/pcre2test)
==36076== Address 0x524b490 is 0 bytes after a block of size 16,384 alloc'd
==36076== at 0x4C2FA3F: malloc (in
/usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==36076== by 0x4C31D84: realloc (in
/usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==36076== by 0x110283: process_data (in pcre-10.34-RC1/pcre2test)
==36076== by 0x115703: main (in pcre-10.34-RC1/pcre2test)
==36076==
==36076== ERROR SUMMARY: 1 errors from 1 contexts (suppressed: 0 from 0)
====================
Crash triage:
--------------------
Function: GETCHARLEN(d, Feptr, len);
// Feptr = (PCRE2_SPTR8) 0x6290000041fb "\377\f00}"
// 0x6290000041fb: 0xff 0x0c 0x30 0x30 0x7d 0x00
GETCHARLEN(d, Feptr, len){
d = *F->eptr;
if (d >= 0xc0u) { //executes this condition d = 0xff
if ((d & 0x20u) == 0) {
d = ((d & 0x1fu) << 6) | (F->eptr[1] & 0x3fu);
len++;
} else if ((d & 0x10u) == 0) {
d = ((d & 0x0fu) << 12) | ((F->eptr[1] & 0x3fu) << 6) |
(F->eptr[2] & 0x3fu);
len += 2;
} else if ((d & 0x08u) == 0) {
d = ((d & 0x07u) << 18) | ((F->eptr[1] & 0x3fu) << 12) |
((F->eptr[2] & 0x3fu) << 6) | (F->eptr[3] & 0x3fu);
len += 3;
} else if ((d & 0x04u) == 0) {
d = ((d & 0x03u) << 24) | ((F->eptr[1] & 0x3fu) << 18) |
((F->eptr[2] & 0x3fu) << 12) | ((F->eptr[3] & 0x3fu) << 6) | (F->eptr[4] &
0x3fu);
len += 4;
} else { //executes this block
d = ((d & 0x01u) << 30) | ((F->eptr[1] & 0x3fu) << 24) |
((F->eptr[2] & 0x3fu) << 18) | ((F->eptr[3] & 0x3fu) << 12) | ((F->eptr[4] &
0x3fu) << 6) | (F->eptr[5] & 0x3fu); //F->eptr[5] leads to a heap
overflow, does not check the length of F->eptr
len += 5;
}
}
}
=====================
--
You are receiving this mail because:
You are on the CC list for the bug.
